Stability analysis and enhancement for the power conversion system
This thesis investigates the instability problems of the ac/dc power conversion system due to the interactions between the EMI filter and the three-phase interlinking converter. A virtual impedance regulator (VIR) is proposed by this thesis to improve stability based on the detailed model of the pow...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Thesis-Doctor of Philosophy |
Language: | English |
Published: |
Nanyang Technological University
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/154687 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | This thesis investigates the instability problems of the ac/dc power conversion system due to the interactions between the EMI filter and the three-phase interlinking converter. A virtual impedance regulator (VIR) is proposed by this thesis to improve stability based on the detailed model of the power conversion system. This proposed virtual impedance can shape the input impedance of the source subsystem only in the specific frequency range. Therefore, the system can be stabilized while ensuring dynamic performance. The power conversion systems in grid-connected and stand-alone operation are both detailed discussed with the proposed VIR. The robust design of the virtual impedance regulator considering the deviation of the system model parameters is also proposed and verified. Then, the dichotomy-based model predictive control (DS-MPC) method with a fixed switching frequency is also proposed for the converter in the weak grid. The proposed fast dichotomy algorithm can select and deduce the optimal voltage vector dynamically through the space vector plane. Therefore, the proposed DS-MPC strategy could ensure dynamic performance and steady-state performance as well. Finally, experimental results are provided to validate the effectiveness of the proposed virtual impedance regulator. |
---|